English

Tokenizing Numerical and Embedding Features for LLM RecSys

Information Retrieval 2026-07-10 v1 Machine Learning

Abstract

Large language models (LLMs) are increasingly used as backbone architectures for recommender systems because of their strong sequence modeling and representation learning capabilities. However, most LLM-based recommenders operate primarily on discrete textual tokens, whereas practical recommendation pipelines also rely on continuous numerical features and dense embedding features produced by upstream feature engineering or pretrained encoders. This mismatch limits the ability of LLM-based models to exploit fine-grained non-textual signals. We propose a soft-token fusion framework that maps numerical and embedding features into the LLM embedding space, allowing heterogeneous recommendation signals to be consumed through the standard token interface. We instantiate the framework in a shared-parameter LLM-based two-tower retrieval model and introduce an interaction-based fusion module that refines embedding and numerical soft tokens before they are inserted into the final LLM input. Experiments on three Amazon recommendation benchmarks show that soft-token fusion improves retrieval performance over LLM-based baselines, and that interaction-based fusion is more effective than direct concatenation of heterogeneous soft tokens.

Cite

@article{arxiv.2607.10016,
  title  = {Tokenizing Numerical and Embedding Features for LLM RecSys},
  author = {Zhe Xu and Ankit Peshin and Chiyu Zhang and Feng Qi and Johnson Lui and Anil Ramakrishna and Justin Johnson and Carl Hu and Kaushik Rangadurai and Luke Simon},
  journal= {arXiv preprint arXiv:2607.10016},
  year   = {2026}
}